Re: [evlog-dev] kernel register facility test

James Keniston <[email protected]> Tue, 8 Jun 2004 09:57:01 -0700
Newsgroups gmane.linux.kernel.event-logging
Message-ID <OF437979CD.F45C3DAD-ON88256EAD.005B2CF8-88256EAD.005D1C75@us.ibm.com>
--0__=07BBE43EDFC8AA688f9e8a93df938690918c07BBE43EDFC8AA68
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: quoted-printable






[email protected] wrote on 06/08/2004 12:44:=
04
AM:

>
> Hi Hien,
>
> I run the kernel test3.sh under cvs kernel/test, I got below error
> message. What should be the problem? I suspect kfacreg should use
> LOGMGMT, but I cannot find the code.
>
> # ./test3.sh
> 1,3c1,8
> < facility=3D0xd8c0bf86 event_type=3D100
> < Registration of facility "test_kfacreg" succeeded.  Facility code =3D=

> 0xf40b2da0.
> < facility=3D0xd8c0bf86 event_type=3D102
> ---
> > facility=3DLOGMGMT event_type=3D40
> > Request to register facility with name =3D test_kfacreg
> > facility code used by kernel =3D 0xd8c0bf86
> > facility code stored in registry =3D 0xd8c0bf86
> > request status =3D OK
> > facility=3Dtest_kfacreg event_type=3D100
> > Registration of facility "test_kfacreg" succeeded.  Facility code =3D=

> 0xd8c0bf86.
> > facility=3Dtest_kfacreg event_type=3D102
> Kernel facility registration test 3 :FAILED
>
>
> Thanks,
> -yi
> ...

test3.sh and test4.sh should be removed from the kernel test suite.  Th=
ey
test the kernel's evl_register_facility() function, which is essentiall=
y a
no-op in v2.6.* versions of the evlog kernel patch.

Here was the thinking: The original implementation of the underlying
evl_gen_facility_code() function is no longer needed, because now, in t=
he
kernel, the facility "code" is just the facility name.  Similarly, the
evl_register_facility() function doesn't do anything that you can't do =
with
the evlfacility -a command.  We included stub versions of these functio=
ns
in the legacy-2.6.* patches to support existing code.
evl_register_facility() returns 0 (success), but doesn't actually log t=
he
event that caused evlogd to register the facility.  We reduced the size=
 of
the evlog patches quite a bit by removing the guts of these functions.

See evlog's legacy-2.6.4.patch for details.

Sorry for any confusion.

Jim Keniston
IBM LTC-RAS

> _______________________________________________
> evlog-developers mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/evlog-developers=

--0__=07BBE43EDFC8AA688f9e8a93df938690918c07BBE43EDFC8AA68
Content-type: text/html; charset=US-ASCII
Content-Disposition: inline
Content-transfer-encoding: quoted-printable

<html><body>
<p><tt>[email protected] wrote on 06/08/2004=
 12:44:04 AM:<br>
<br>
&gt; <br>
&gt; Hi Hien,<br>
&gt; <br>
&gt; I run the kernel test3.sh under cvs kernel/test, I got below error=
<br>
&gt; message. What should be the problem? I suspect kfacreg should use<=
br>
&gt; LOGMGMT, but I cannot find the code.<br>
&gt; <br>
&gt; # ./test3.sh<br>
&gt; 1,3c1,8<br>
&gt; &lt; facility=3D0xd8c0bf86 event_type=3D100<br>
&gt; &lt; Registration of facility &quot;test_kfacreg&quot; succeeded. =
&nbsp;Facility code =3D<br>
&gt; 0xf40b2da0.<br>
&gt; &lt; facility=3D0xd8c0bf86 event_type=3D102<br>
&gt; ---<br>
&gt; &gt; facility=3DLOGMGMT event_type=3D40<br>
&gt; &gt; Request to register facility with name =3D test_kfacreg<br>
&gt; &gt; facility code used by kernel =3D 0xd8c0bf86<br>
&gt; &gt; facility code stored in registry =3D 0xd8c0bf86<br>
&gt; &gt; request status =3D OK<br>
&gt; &gt; facility=3Dtest_kfacreg event_type=3D100<br>
&gt; &gt; Registration of facility &quot;test_kfacreg&quot; succeeded. =
&nbsp;Facility code =3D<br>
&gt; 0xd8c0bf86.<br>
&gt; &gt; facility=3Dtest_kfacreg event_type=3D102<br>
&gt; Kernel facility registration test 3 :FAILED<br>
&gt; <br>
&gt; <br>
&gt; Thanks,<br>
&gt; -yi<br>
&gt; ...</tt><br>
<br>
<tt>test3.sh and test4.sh should be removed from the kernel test suite.=
 &nbsp;They test the kernel's </tt><tt>evl_register_facility() function=
, which is essentially a no-op in v2.6.* versions of the evlog kernel p=
atch.</tt><br>
<br>
<tt>Here was the thinking: The original implementation of the underlyin=
g evl_gen_facility_code() function is no longer needed, because now, in=
 the kernel, the facility &quot;code&quot; is just the facility name. &=
nbsp;Similarly, the evl_register_facility() function doesn't do anythin=
g that you can't do with the evlfacility -a command. &nbsp;We included =
stub versions of these functions in the legacy-2.6.* patches to support=
 existing code. &nbsp;evl_register_facility() returns 0 (success), but =
doesn't actually log the event that caused evlogd to register the facil=
ity. &nbsp;We reduced the size of the evlog patches quite a bit by remo=
ving the guts of these functions.</tt><br>
<br>
<tt>See evlog's </tt><tt>legacy-2.6.4.patch for details.</tt><br>
<br>
<tt>Sorry for any confusion.</tt><br>
<br>
<tt>Jim Keniston</tt><br>
<tt>IBM LTC-RAS</tt><br>
<tt><br>
&gt; _______________________________________________<br>
&gt; evlog-developers mailing list<br>
&gt; [email protected]<br>
&gt; <a href=3D"https://lists.sourceforge.net/lists/listinfo/evlog-deve=
lopers">https://lists.sourceforge.net/lists/listinfo/evlog-developers</=
a><br>
</tt></body></html>=

--0__=07BBE43EDFC8AA688f9e8a93df938690918c07BBE43EDFC8AA68--



-------------------------------------------------------
This SF.Net email is sponsored by: GNOME Foundation
Hackers Unite!  GUADEC: The world's #1 Open Source Desktop Event.
GNOME Users and Developers European Conference, 28-30th June in Norway
http://2004/guadec.org